-
@strengejacke As you can see, it differs from broom by giving more flexibility, computing more things, and also being less "generic" (e.g. the colnames directly refers to what they represent, i.e., Me…
-
(suggested by @avehtari)
Now that we have basic k-fold cross-validation in **rstanarm** we should implement stratified and grouped K-fold-CV.
-
Add a feature to module search to allow using the dependencies from an easyconfig. Example: add tidymodels to an existing R easyconfig. Searching for the module tidymodels produces 139 dependencies.…
-
Because in both, `brms` and `rstanarm`, the R^2 metric is not available for ordinal models (in contrast with elpd, loo or waic), `model_performance(ordinal_model )` does not work.
-
Let's do this for rstanarm
https://groups.google.com/d/msg/stan-users/5IxMxY3tv2w/ptm2vyJmBwAJ
see the repo https://github.com/jtpi/rstan-varsel
-
#### Summary:
The `s` function in stan_gamm4 throws an error when bs = "cc".
#### Description:
Despite working until relatively recently, use of the `s` function now appears to fail when trying …
-
- [x] ci
- [x] equivalence_test
- [x] estimate_density
- [x] hdi
- [x] rope
- [x] p_direction
> This would also mean that the plotting functions in see need to be updated to use as.data.frame(…
-
Should we add covar(intercepts, slopes) to OSPREE models?
-
Something like `ggthemes::theme_solarized(light = FALSE)`, but with a more contrastful color for axis labels, and maybe more beatiful legend?
Here's the ggthemes-theme:
``` r
library(easystats)…
-
``` r
library(rstanarm)
library(emmeans)
library(bayestestR)
library(see)
fit Warning in max(x): no non-missing arguments to max; returning -Inf
#> Warning in min(x): no non-missing arguments…